I was able to reproduce a reference leak with my instrumented code
(plus my regular set of network patches, so problem could still be me).
The reference lost was acquired here:
struct neighbour *neigh_create(struct neigh_table *tbl, const void *pkey,
struct net_dev
On Sun, Aug 28, 2005 at 07:26:12PM -0400, Jeff Garzik wrote:
> applied, but let us know when the root cause is found...
local->hw_priv was initialized only after the interrupt handler was
registered. This could trigger a NULL pointer dereference in
prism2_pccard_card_present() that assumed that lo
applied, but let us know when the root cause is found...
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
applied patches 1-5
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi,
I noticed a discussion on netdev
with someone getting dst cache overflow errors; I've been experiencing
similar things on a router - it has died 3 or 4 times (over a period
of a few months) with such an error with very little traffic passing
through it and a stream of the 'dst cache overflow'
jamal wrote:
On Sun, 2005-28-08 at 01:19 -0700, Ben Greear wrote:
Here's an updated patch (git working better now, thanks!):
http://www.candelatech.com/oss/rfcnt.patch
I started a list of constants #defined in netdevice.h. These
will be used for all stack variable references and other place
With my Buffalo WLI-CF-S11G PC Card kernel oopses every time in
prism2_interrupt() when I try load the hostap module. local->hw_priv is null
during the first call to prism2_interrupt(). It feels like
interrupts are enabled too early, or something.
This patch fixes the symptom, but not the cause.
hostap_cs: 0.4.1-kernel (Jouni Malinen <[EMAIL PROTECTED]>)
pcmcia: hostap_cs: invalid hash for product string "BUFFALO": is 0x1b01a57b,
should be 0x2decece3
pcmcia: see Documentation/pcmcia/devicetable.txt for details
pcmcia: hostap_cs: invalid hash for product string "WLI-CF-S11G": is
0xefd5102a,
The first three patches are cleanup for ieee80211 (netdev-2.6, ieee80211
branch) and the last two are for hostap (netdev-2.6, ieee80211-wifi
branch). Please apply to these branches.
--
Jouni MalinenPGP id EFC895FA
-
To unsubscribe from this list: send t
Version 0.4.4 of Host AP driver was released, so let's sync the version
number in netdev-2.6 tree.
Signed-off-by: Jouni Malinen <[EMAIL PROTECTED]>
Index: netdev-2.6/drivers/net/wireless/hostap/hostap_config.h
===
--- netdev-2.6.orig
IEEE 802.11 code has no business touching payloads of EAPOL frames.
There are some EAPOL structures defined for debugging and these were
confusingly called EAP types which they are not. Let's just remove these
before someone else starts using them in the kernel.
Signed-off-by: Jouni Malinen <[EMAI
No need to maintain support for WIRELESS_EXT < 17 since this kernel
tree is already using WIRELESS_EXT 18.
Signed-off-by: Jouni Malinen <[EMAIL PROTECTED]>
Index: netdev-2.6/include/net/ieee80211.h
===
--- netdev-2.6.orig/include/net
Debug variables and procfs dir should be "ieee80211", not "ipw".
Signed-off-by: Jouni Malinen <[EMAIL PROTECTED]>
Index: netdev-2.6/include/net/ieee80211.h
===
--- netdev-2.6.orig/include/net/ieee80211.h
+++ netdev-2.6/include/net/ie
On Sun, 2005-28-08 at 01:19 -0700, Ben Greear wrote:
> Here's an updated patch (git working better now, thanks!):
>
> http://www.candelatech.com/oss/rfcnt.patch
>
> I started a list of constants #defined in netdevice.h. These
> will be used for all stack variable references and other places
> w
On Sun, 2005-28-08 at 10:13 +0200, Eric Dumazet wrote:
> traffic is about 50 Mb/s, 35000 packets sent per second, so this is small
> compared to the max specs of this hardware.
>
That doesnt look bad at all. It cant be the bus (which is typically
where you see a lot of those requeues). I think
On 8/27/05, Eric Lemoine <[EMAIL PROTECTED]> wrote:
> On 8/27/05, Eric Lemoine <[EMAIL PROTECTED]> wrote:
> > On 8/26/05, Geoff Levand <[EMAIL PROTECTED]> wrote:
> > > This fixes a major bug in the Sun GEM Ether
> > > driver's netpoll implementation. When both polled
> > > and interrupt driven i/o
David S. Miller wrote:
From: Ben Greear <[EMAIL PROTECTED]>
Date: Sat, 27 Aug 2005 11:19:24 -0700
Although my implementation is not there yet, it would be nice if we
could use the address of the pointer than holds the dev reference
as the key. We should then be able to also catch double deref
jamal a écrit :
On Sat, 2005-27-08 at 22:38 +0200, Eric Dumazet wrote:
(So about 360 requeues per second, much better than before (12000 / second))
I suspect what you are doing is shoving a lot more packets than the wire
can handle. Thats why you are getting the backpressure.
I read back
18 matches
Mail list logo
